Abstract 4370204: Incidence of Myocarditis in Patients Receiving Immune Checkpoint Inhibitors

Abstract

Background: Myocarditis has been recognized as an uncommon but potentially severe side effect of immune checkpoint inhibitors (ICIs). Single-center studies utilizing high-quality chart review have demonstrated incidence of approximately 1-2% in patients receiving ICIs; however, claims-based analyses have generally shown lower rates. This study was intended to evaluate the rate of myocarditis diagnosis in a national electronic health record (EHR) dataset. Research Question: What is the incidence of myocarditis among patients receiving ICIs in a national EHR dataset? Methods: This study was deemed exempt from review by the Yale University IRB as non-human subject research due to deidentification. Data used in this study came from Epic Cosmos, a dataset created in collaboration with a community of Epic health systems representing more than 300 million patient records from over 1,744 hospitals and 40,700 clinics. Analysis was limited to sites in the United States from January 1, 2014, to December 31, 2024. ICI exposure was defined as clinic or hospital administration of an ICI. Only ICIs with greater than 10,000 exposures were included. Incident myocarditis was defined as an encounter-level diagnosis of ICD-10-CM I51.4 or I40* after first ICI administration. Data analysis was performed in R (v4.2.2, R Core Team). Results: A total of 397,753 ICI exposures were observed. At 36 months, 0.35% (95%CI 0.33-0.37%) of ICI exposures developed myocarditis with 62% of cases occurred within the first month. The highest overall incidence of myocarditis was with ipilimumab, 0.67% (95%CI 0.59-0.76%). The lowest incidence of myocarditis was with atezolizumab, 0.17% (95%CI 0.13-0.22%). Trends by agent are shown in Figure 1. Conclusions: Myocarditis encounter diagnoses were observed in 0.35% of ICI exposures by 36 months. Most of these diagnoses occurred within the first month of therapy. Further analysis of the patient-level predictors of myocarditis in this national dataset is warranted.